Inducing Optimal Attribute Representations for Conditional GANs
This work solves the problem of generating more natural and attribute-clear synthetic images for researchers and practitioners in computer vision, though it is incremental as it builds on existing cGAN architectures.
The paper tackles the problem of improving conditional GANs for image translation by addressing the limitations of hard-coded categorical vectors, proposing a novel framework using Graph Convolutional Networks to learn attribute representations. The result shows that enhanced cGANs outperform state-of-the-art methods by a large margin on datasets like CelebA, with improvements in attributes recognition rates and quality measures such as PSNR and SSIM.
Conditional GANs are widely used in translating an image from one category to another. Meaningful conditions to GANs provide greater flexibility and control over the nature of the target domain synthetic data. Existing conditional GANs commonly encode target domain label information as hard-coded categorical vectors in the form of 0s and 1s. The major drawbacks of such representations are inability to encode the high-order semantic information of target categories and their relative dependencies. We propose a novel end-to-end learning framework with Graph Convolutional Networks to learn the attribute representations to condition on the generator. The GAN losses, i.e. the discriminator and attribute classification losses, are fed back to the Graph resulting in the synthetic images that are more natural and clearer in attributes. Moreover, prior-arts are given priorities to condition on the generator side, not on the discriminator side of GANs. We apply the conditions to the discriminator side as well via multi-task learning. We enhanced the four state-of-the art cGANs architectures: Stargan, Stargan-JNT, AttGAN and STGAN. Our extensive qualitative and quantitative evaluations on challenging face attributes manipulation data set, CelebA, LFWA, and RaFD, show that the cGANs enhanced by our methods outperform by a large margin, compared to their counter-parts and other conditioning methods, in terms of both target attributes recognition rates and quality measures such as PSNR and SSIM.